metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: source_file_id
dtype: string
- name: segment_id
dtype: string
- name: duration
dtype: float64
- name: speaker_id
dtype: string
- name: gender
dtype: string
- name: dialect
dtype: string
- name: orthography
dtype: string
- name: source_type
dtype: string
splits:
- name: train
num_bytes: 4073006494.408
num_examples: 9691
- name: test
num_bytes: 536288649.075
num_examples: 1195
- name: validation
num_bytes: 522842821.816
num_examples: 1194
download_size: 5340584489
dataset_size: 5132137965.299
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*